docs(tests): tighten the protocols test prose from the post-sweep merges (#719)

[JarryShaw]

• Searched for similar pull requests
• Followed the coding style (make pylint, make mypy, make isort) — not run; comments/docstrings only
• make test passes, and a test case covers the change — each touched module run alone, plus tests/project
• Added a changelog entry — N/A — centralised in docs(changelog): shared 1.5.0 changelog — long-lived, merges last (#610, #616, #617, #618, #620) #657

What is the purpose of your pull request?

• docs — documentation only

---

Description

Final #719 pass over tests/protocols/** prose touched by ef0964c81..efc9c3774. Most of those 50 files only swapped purge_modules for reimport_once_per_class, with no prose. Changes:

• test_schema_metaclass_abc_cache_unit.py: the setUp comment credited purge_modules, which setUp no longer calls. It now names reimport_once_per_class for the cache reset, and says classes stay fresh per test only because tearDown purges before the class import is kept. Verified: the schema module id differs in every test.
• test_dispatch_reachability_unit.py: drop the "no longer one import per test" history.
• test_option_roundtrip_unit.py, test_dispatch_registry_unit.py: the comment cited a setUp purge, but neither class has a setUp; the purge is in setUpClass.
• test_ipv6_ext_unit.py, test_pcapng_unit.py: tightened wording.

Tests: 6 modules pass; tests/project 375 passed, 1 skipped.

File	AST equal (docstrings stripped)
tests/protocols/internet/test_ipv6_ext_unit.py	True
tests/protocols/misc/test_pcapng_unit.py	True
tests/protocols/schema/test_schema_metaclass_abc_cache_unit.py	True
tests/protocols/test_dispatch_reachability_unit.py	True
tests/protocols/test_dispatch_registry_unit.py	True
tests/protocols/test_option_roundtrip_unit.py	True

[github-actions]

Coverage: 88.74% (unit tier, Python 3.14, f010f266a, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	4	93.55%
pcapkit/const	18757	1035	2342	846	90.23%
pcapkit/corekit	1874	91	578	22	94.33%
pcapkit/dumpkit	136	0	40	0	100.00%
pcapkit/foundation	2422	143	842	34	92.62%
pcapkit/interface	112	7	40	5	92.11%
pcapkit/protocols	15653	187	3942	162	98.19%
pcapkit/toolkit	487	71	144	3	84.15%
pcapkit/utilities	429	4	122	4	98.55%
pcapkit/vendor	4409	2359	1006	158	42.84%

Per-file detail: the coverage-html artifact of this run.

[JarryShaw]

Cross-review verdict on f010f266a: GOOD TO GO (ran on Sonnet; author Opus)

• No code changed. All 6 files are AST-identical to origin/main once bare strings are stripped.
• abc-cache setUp comment holds under pytest with conftest active:
  • On the first test, tearDown purges before _keep_lazy_imports runs, so the class holds {}.
  • Later tests restore that empty import, which resets the abc cache.
  • The reviewer's probe saw a distinct Schema object in each of the 5 tests that import it.
• No setUp in the two other classes. Neither has one, here or on main; the purge is in setUpClass, ahead of the generator load.
• Rationale kept. The gc reasoning is unchanged. _sweep() has 4 callers, so "the other three" is right.
• Tests pass. The touched modules and tests/project (375 passed, 1 skipped).
• UNVERIFIED by the reviewer: a full test_pcapng_unit.py run. Only the touched class ran (10 passed). The author's run of the full module gave 97 passed.